Primary infection with varicella-zoster virus (VZV) causes chickenpox, and reactivation of the virus from latency results in zoster. The purpose of this project is to study the molecular pathogenesis and latency of VZV and to identify receptors that the virus uses to infect cells. [unreadable] [unreadable] The major VZV gene expressed during latency is gene 63. Gene 63 RNA and protein have been detected in latently infected human and animal ganglia. VZV deleted for gene 63 is impaired for replication in cell culture and for latency in rodents. Interferon-alpha partially inhibits the growth of VZV in cell culture and in the skin of rodents in an animal model of VZV. Growth of VZV deleted for gene 63 was severely inhibited in the presence interferon-alpha. In contrast, other VZV mutants that are similarly impaired for growth in the absence interferon-alpha were not severely inhibited in the presence of interferon-alpha. The VZV gene 63 deletion mutant was not severely inhibited by interferon-gamma. Iinterferon-alpha inhibited expression of viral proteins, but not viral mRNA, in cells infected with the gene 63 deletion mutant at a post-transcriptional level. Interferon-alpha stimulates gene products that can phosphorylate a protein involved in synthesis of proteins (eIF-2alpha) and inhibit protein synthesis. We found that gene 63 was important for inhibiting phosphorylation of the protein (eIF-2alpha). Since interferon-alpha has previously been shown to limit the growth of VZV in human skin when engrafted onto mice, the ability of VZV gene 63 to block the effects of interferon alpha may play a critical role in VZV pathogenesis. [unreadable] [unreadable] We have previously identified a cellular protein, insulin-degrading enzyme (IDE), that interacts with a VZV glycoprotein, gE. While IDE is predominantly present inside the cells, we and others have shown that a portion of IDE is present on the surface of cells. We further confirmed that IDE interacts with VZV in virus-infected cells. Treatment of cells with several compounds, that inhibit the activity of IDE (including the antibiotic bacitracin) inhibited infection with cell-free virus and reduced cell-to-cell spread of virus. [unreadable] [unreadable] We have mapped the region of gE that is important for its interaction with IDE and the region of IDE important for its interaction with gE. We found that amino acids 24-71 of gE are required for IDE binding. Deletion of amino acids 32-71 of gE resulted in loss of its ability to bind IDE and a peptide corresponding to amino acids 24-50 of gE blocked its interaction with IDE. We also found that the conformation of gE was important for its binding to IDE. A molecule consisting of amino acids 1-71 of VZV gE fused to most of the herpes simplex virus gE did not show an increased level of binding to IDE compared with full length herpes simplex virus gE. VZV gE also forms a complex with VZV glycoprotein, gI. We found that amino acids 163-208 of gE are required for its ability to complex with gI. Therefore, distinct motifs of VZV gE are important for binding to IDE or to gI. Finally we found that gE binds predominantly to the first half of the IDE molecule.[unreadable] [unreadable] VZV gene 29 is expressed during latency in human ganglia. We constructed a VZV mutant deleted for gene 29 and found that the gene is required for growth in cell culture. We also made a virus which lacks the normal copy of gene 29, but contains another copy of gene 29 whose expression is regulated by a different viral gene element (the cytomegalovirus promoter), and therefore overexpresses the 29 gene. This latter virus grew similar to the wild-type virus in cell culture, but surprising was impaired for latency in a rodent model of VZV infection. These studies indicate that regulation of gene 29 at appropriate levels is critical for VZV latency in a rodent model. Since the current chickenpox vaccine occasionally reactivates from latency and cause shingles, the virus which overexpresses gene 29 and is impaired for latency might be less likely to reactivate and cause shingles, and therefore might have features that would be desirable for a safer chickenpox vaccine.